Impaired arbitration between reward-related decision-making strategies in Alcohol Users compared to Alcohol...

Area of Science:

• Neuroscience
• Cognitive Psychology
• Addiction Research

Background:

• Decision-making involves a balance between habitual (model-free) and goal-directed (model-based) control.
• Model-based control is typically enhanced in high-reward situations to optimize outcomes.
• Frequent alcohol use may affect reward sensitivity and decision-making strategies.

Purpose of the Study:

• To investigate how frequent alcohol users arbitrate between model-free and model-based control under varying reward magnitudes.
• To assess risk sensitivity in alcohol users and non-users across different reward stakes.
• To explore the relationship between model-based control and risk-taking behavior in frequent alcohol users.

Main Methods:

• A modified 2-step reinforcement learning task with low and high reward conditions was administered to 81 participants (47 alcohol users, 34 non-users).
• A dual-system reinforcement learning model (maximum a posteriori fitting) quantified model-based control.
• A utility model assessed risk sensitivity for low- and high-stakes trials.

Main Results:

• Alcohol non-users exhibited increased model-based control in high-reward vs. low-reward conditions; alcohol users did not show this difference.
• Both groups were less risk-averse in high-reward conditions.
• Alcohol users were significantly less risk-averse than non-users in high-reward conditions.
• Greater model-based control correlated with less risk-sensitivity in alcohol users.

Conclusions:

• Frequent alcohol users demonstrate less flexible decision-making, failing to adapt control strategies based on reward magnitude.
• Alcohol users exhibit heightened risk-proneness, particularly in high-stakes scenarios.
• Findings suggest potential impairments in metacontrol among frequent alcohol users, impacting adaptive behavior in response to reward variability.